What is physiological gastroesophageal reflux?
J.R. Bennett (Hull)
This paper will discuss the phenomenon of gastroesophageal reflux disease (GERD) which is not accompanied by symptoms nor endoscopic abnormalities and does not lead to complications. It is at present detectable only by intraesophageal pH monitoring. There are many technical aspects of pH monitoring which affect the results of such tests but they will not be dealt with except where variations of methodology cause registrable acid reflux which would not be revealed by other techniques.
pH recordings in "normal" individuals
Although esophageal pH had been measured over short periods since the 1950s, it was not until the 1960s that prolonged recordings were made. It was immediately seen [1-3] that "normal" subjects had short episodes of acid reflux mainly after meals but uncommonly at night.
This has since been repeatedly observed  but it poses a question about the selection of "normal" subjects. Careful epidemiological studies show that only one-third of British subjects never experience heartburn . Few studies have been as careful as that of Johnsson et al. (1987)  whose normal subjects had "no past or present symptoms of gastrointestinal disease... " and all had normal endoscopic
examination (9% of their asymptomatic subjects had an endoscopic abnormality and were rejected). They also pointed out that the resulting pH data was skewed, and so used the 95th percentile as the upper reference limit. Having obtained clear discrimination between 20 normal and 20 symptomatic patients the study was enlarged. Overlap then resulted, four patients having results within the normal range and one normal subject being above it. All other studies have also shown an overlap between their symptomatic patients and normal controls, however, the latter were selected [7-12].
Moreover, Smout found a correlation between age and duration of acid reflux, so that in subjects over 45 the 95th percentile figure for % time below 4 was 12, which was much higher than generally recognised "normal values". However, Kruse-Andersen et al.  did not find any increase in reflux values when 10 healthy subjects were studied 8 years apart. Alternative analysis using receiver operator characteristics (ROC) to identify the best discriminatory values  indicate pH 4 as the best cut-off point and demonstrate threshold values similar to those derived by standard statistical methods. Nevertheless, overlap between values for normal and symptomatic subjects remains.
It is therefore apparent that in prolonged recording of the lower esophageal pH, falls below pH 4, presumptively indicating acid gastresophageal reflux, can be detected in almost every asymptomatic, endoscopically normal person. Furthermore, in some of these apparently normal individuals the frequency and duration of such episodes can be greater than in some patients with symptoms attributable to reflux or with endoscopic changes of reflux esophagitis, even when matched for age. Some subjects (especially older ones) may have values for pH exposure well into the "pathological" range.
pH recording related to symptoms
A study of 86 patients with "symptoms typical of gastroesophageal reflux" graded by severity showed the usual correlation between severity of symptoms and mean pH reflux time, but 18 were within the normal range including four with the most severe symptoms .
Twelve patients with endoscopic esophagitis also fell into the normal range of pH recording, three of them with confluent erosions, ulcer/stenosis or Barrett's.
When 304 patients had pH monitoring for suspected GERD, heartburn occurred in 48% of the 138 with normal pH results compared to 68% of them with abnormal reflux.
Schlesinger and colleagues  found that of 30 patients with "reflux symptoms" and a normal endoscopy, only 21% had pH reflux outside normal values (although the group as a whole were significantly greater than normal). Ninety-three percent of those with normal studies responded to antireflux therapy.
Correlation between pH studies and endoscopy show a limited concordance of only 60% .
Reflux episodes related to symptoms
The studies above relate the sum of reflux episodes in a given period (usually 24 h) to symptoms extending over longer periods (days, months). There have been attempts to relate specific symptomatic episodes to single reflux events. The symptom usually investigated has been chest pain rather than more characteristic heartburn. The frequency of a close correlation of reflux with pain has varied from 24 to 62% [16-18] but it is not stated how many pH recorded reflux events were painless. Ghillebert et al. (1990) , used a binomial formula to indicate a probability of pain associated with reflux being due to chance, while Armstrong and colleagues (1992)  developed a computer program which compared the frequency distributions of pH events associated with pain, with the frequency distribution of symptom-unrelated pH events.
Acid sensitivity of the esophagus has traditionally been determined by standardised acid perfusion of the lower esophagus ("Bernstein test"). Correlation between the results of this test and scores for pain related to acid reflux during ambulatory monitoring, are poor.
In Hewson and colleagues (1989)  study, only 48% of Bernstein-positive patients had acid-related pain by pH monitoring; Kruse-Andersen (personal communication) found no difference in the frequency of pain episodes related to reflux between patients who were acid perfusion positive or negative.
Intraesophageal pH is a continuous variable in the community, with a distribution skewed to the right. The greater the acid exposure of the esophagus, the more likely the subject is to have reflux symptoms or esophagitis, but other factors must be important in determining these consequences. The corollary of this is that it is simplistic to define results of pH recording as being "normal" or "abnormal". The quantitative result of such recordings must be considered as one piece of evidence to be analysed in combination with symptoms and endoscopic changes; where possible, a score relating reflux episodes to symptoms is desirable.
8. Vitale GC, Cheadle WG, Gadek S, Michel ME, Cuschieri A. Computerized 24-hour ambulatory esophageal pH-monitoring and esophageal gastro-duodenoscopy in the reflux patient. Ann Surg 1984;200:724-729.
9. Schindlbeck NE, Heinrich C, Konig A, Dendorfer A, Pace F, Muller-Lissner A. Optimal thresholds, sensitivity and specificity of long term pH monitoring for the detection of gastroesophageal reflux disease. Gastroenterology I987;93: 85-90.
10. Mattioli V, Pilotti V, Zannoli R, Carvui N, Loria P, Felice V, Conci A, Castellini A, Gozzetti G. Twenty-four Hour Home Esophago-gastro pH Monitoring: Standardisation of the Method. Third European Symposium on Gastrointestinal Motility, Bruges, 1986:91.
12. Smout AJPM, Breedijk M, van der Zouw C, Akkermans LMA. Physiological gastroesophageal reflux and esophageal motor activity studied with a new system for 24-hour recording and automated analysis. Dig Dis Sci 1989;34:372-378.
17. DeMeester TR, O'Sullivan GC, Bermudez CA, Midell A, Cimochowski GE, O'Drobinak J. Esophageal function in patients with angina-type chest pain and normal coronary angiogram. Ann Surg 1982:4:488-498.
19. Ghillebert G, Janssens J, Vantrappen G, Nevens F, Piessens J. Ambulatory 24-hour intraoesophageal pH and pressure recordings vs. provocation tests in the diagnosis of chest pain of oesophageal origin. Gut 1990;31:738-744
J. Janssens (Leuven)
Brief episodes of acid reflux occur in healthy subjects, especially in the postprandial period. During the day the intraesophageal pH is less than 4, for about 2% of the recording time; nocturnal episodes of reflux are normally very rare. These short-lived episodes of reflux remain asymptomatic and do not result in esophagitis; they correspond to what is called physiological reflux.
Symptoms and lesions of esophagitis appear when the episodes of reflux become more frequent and more prolonged; the reflux is then considered to be pathological. In normal subjects, as in patients with pathological reflux, nearly all the episodes of reflux obey one of the following three mechanisms:
- total transient relaxation of the gastroesophageal sphincter;
- a transient increase in intra-abdominal pressure;
- spontaneous reflux through a very hypotonic sphincter.
Physiological reflux nearly always follows transient relaxation of the lower
esophageal sphincter (LES) (94% of cases); while in patients with esophagitis this
mechanism is responsible for only 65% of the episodes of reflux. A transient increase in intra-abdominal pressure and spontaneous reflux are responsible for 17 and 18% of episodes of reflux respectively in patients with esophagitis .
The mechanism responsible for the induction of transient lower esophageal sphincter (tLES) relaxations is unclear. Some authors believe that transient LES relaxations are triggered by partial or incomplete swallowing [2,3]. Others feel that gastric distension is the major factor that induces the transient LES relaxations via vagovagal reflexes initiated by stimulation of mechanoreceptors in the gastric wall . This mechanism could also explain the increased rate of transient LES relaxations and reflux during the postprandial period . Transient LES relaxations may also occur at night, but almost exclusively during periods of being awake.
Relatively little is known about the occurrence and control of transient LES relaxations in patients with reflux disease. Dodds and Dent found a higher rate of transient LES relaxations in reflux patients [1,6], while Mittal and McCallum found that normals and patients had the same rate of tLES relaxations, but that the tLES relaxations were more frequently accompanied by reflux in the patient group [2,7]. When an episode of reflux has occurred, it is important that the refluxed material be evacuated from the esophagus as fast as possible. Besides the more or less irritant nature of the refluxed material, it is the duration of contact with the mucosa which determines whether or not esophagitis develops. Esophagitis seems to be more particularly related to long-lasting episodes of reflux.
In the upright position, gravity plays an important role in the elimination of this refluxed material. This is why patients with orthostatic reflux develop esophagitis less frequently than those with decubitus reflux or mixed reflux. In the prone position, (primary or secondary) esophageal peristalsis is mainly responsible for the clearance of acid. Although gravity and peristalsis are capable of returning nearly all the reflux material to the stomach, that in itself does not alter the pH of the tiny amount which still remains in contact with the esophageal mucosa. This residue must be neutralized chemically by saliva . Physiological reflux is rapidly cleared from the lumen: it almost exclusively occurs in the upright position when gravity clears the lumen; if reflux occurs in the prone position normal peristalsis guarantees a fast clearance.
Twenty-four hour intraesophageal pH recording has become a routine procedure in clinical practice. Automatic analysis calculates various reflux parameters (i.e., the number of reflux episodes, the percentage of time of the recording when the pH is below 4, which is an expression of the total exposure of the esophageal mucosa to acid), the mean duration of the reflux episodes and the number of long lasting (> 5 min) reflux episodes (which reflect the efficiency of the acid clearing function). When these variables are determined in a group of healthy controls, they define in another way what is called "physiological reflux" . If one uses the numerical parameters of a 24-h pH recording to distinguish normal subjects from patients with gastroesophageal reflux disease (GERD), the technique is excellent to distinguish normals from patients with erosive esophagitis; however, when patients with symptoms without lesions are included, the distinction is less clear [10,11]. Patients with symptomatic GERD, as a group have on intraesophageal pH measurements an increased exposure to acid as compared to control subjects . Patients with almost continuous
symptoms have a much greater acid exposure than patients with daily symptoms or with occasional symptoms only . However, symptomatic reflux and/or a positive acid perfusion test may be found in patients with normal esophageal acid exposure time during pH monitoring .
Although heartburn is the regular symptom induced by acid reflux, in some patients reflux may be responsible for pseudoanginal chest pain rather than heartburn. Prolonged measurements of esophageal pH has shown that, in subjects with angina-like chest pain of esophageal origin, most of the painful episodes appear to be caused by reflux, with or without a disorder of esophageal motor function . In a recent overview it was shown that of all chest pain episodes that were of proven esophageal origin, as many as 71% of them appeared to be acid related . In some patients with esophageal angina-like chest pain, an identical painful episode may be provoked either by isolated reflux (without motor disorder) or by severe alterations in motor function (without reflux) or again, at other times, by the association of reflux and motor disorders. Esophageal acid perfusion (Bernstein's test) may sometimes provoke this typical painful episode in patients in whom spontaneous episodes of pain seem to be connected solely with motor disorders without reflux. There therefore seems to be a group of patients in whom the esophagus is sensitive to numerous stimuli which are different from each other; it is this entity that has been called "irritable esophagus" .
Numerical analysis of the 24-h pH recording in these patients frequently appears to fall into normal limits.
Physiological reflux should be defined as the amount of reflux which occurs in a normal subject, who does not complain of reflux symptoms and has no endoscopic esophagitis. This physiological reflux can be quantified by numerical analysis of a 24-h pH recording. However, some patients with typical reflux symptoms and especially patients with atypical reflux symptoms (i.e., chest pain) have a numerical analysis of a 24-h pH recording which is in the normal range, although their reflux clearly is not physiological.
14. Howard PJ, Maher L, Pryde A, Heading RC. Symptomatic gastro-oesophageal reflux, abnormal oesophageal acid exposure, and mucosal acid sensitivity are three separate, though related, aspects of gastro-oesophageal reflux disease. Gut 1991; 32:128-132.
R.S. Fisher (Philadelphia)
Gastroesophageal reflux is defined as the retrograde movement of the stomach contents across the gastroesophageal junction into the esophagus. Reflux can be regarded as physiological as long as it does not cause symptoms or damage the esophageal mucosa. In this monograph the composition, measurement and regulation of physiological gastroesophageal reflux will be discussed.
Composition of the refluxate
Over time the composition of the gastroesophageal refluxate reflects the stomach contents. During a 24-h period the stomach contains variable quantities of hydrochloric acid, pepsinogen/pepsin (proteolytic enzyme), mucus, and digestive products derived from food. The stomach contents change continuously depending upon the quantity of acid secreted, the ratio between the inactive pepsinogens and active pepsins, the amount of mucus and bicarbonate secreted by the gastric mucosa, and the composition of ingested meals. In some situations enterogastric reflux may occur introducing bile salts and, perhaps, pancreatic enzymes into the stomach and making them available to regurgitate into the esophagus. Bacterial overgrowth in the stomach or proximal small intestine can alter the digestive products and/or the structure of bile
salts within the stomach. These changes may be important because they could alter the potency of the refluxate.
Gastric acid secretion has been studied basally and in response to meals and secretagogues such as pentagastrin. In addition, 24-h intragastric pH profiles have characterized the effects of different meals, gastric emptying, and sleep. It is important to recognize that there are times during the day when the normal stomach contains little acid. Therefore, gastroesophageal reflux would not be detected by intraesophageal pH monitoring. Of interest, these periods of physiologic gastric alkalinity may occur postprandially when the lower esophageal sphincter (LES) pressure may be reduced by transient lower esophageal sphincter relaxations.
Measurement of gastroesophageal reflux
Much of our knowledge about gastroesophageal reflux is based upon detection of acid within the esophagus using glass or antimony, hydrogen-detecting electrodes placed at varying distances, usually 5 cm, above the LES. Using 24-h intraesophageal pH monitoring, esophageal pH has been reported to be less than 4 for approximately 6% of the total time, 8% of the upright (postprandial) time and 2% of the recumbent (sleeping) time [4,5,25]. These "normal values" are approximations based on numerous reports. Although esophageal pH monitoring may be useful to correlate symptoms with acid reflux episodes into the esophagus and to establish guidelines for acid exposure, this technique has a number of limitations. First, 24-h intraesophageal pH monitoring does not record the volume of reflux, but only measures acid concentrations in the microenvironment of the pH electrode. Secondly, pH monitoring detects acid reflux, but not reflux of bile salts (primary or secondary), pancreatic enzymes, or abnormal digestive products. Also, it is well established that the intragastric pH is not below 4 continuously throughout the day. Therefore, reflux episodes which occur during nonacid periods would not be detected by pH monitoring. Thirdly, esophageal pH monitoring could be affected by salivary volume and composition and by bicarbonate secretion from the esophageal, stomach and duodenal mucosa and from the pancreas.
Gastroesophageal scintigraphy has been employed in some centers to evaluate gastroesophageal reflux . This technique utilizes emitting ע radionuclides mixed with the stomach contents, not only to detect, but also to quantitate reflux. If the radionuclide mixes homogeneously with the gastric contents, counting y scintillations over the esophagus will reflect gastroesophageal reflux of both the acid and nonacid gastric contents. Unfortunately, gastroesophageal scintigraphy is limited by gastric emptying. Once the y labeled gastric contents have emptied from the stomach, this technique will not demonstrate gastroesophageal reflux. Therefore, it cannot be used for prolonged studies beyond 1 or 2 h. Roentgenographic techniques utilizing radiopaque barium have also been used to detect reflux, but are insensitive and non-quantitative.
Whether bile salts, pancreatic enzymes, or high pH (alkaline) liquid regurgitate normally from the duodenum into the stomach and then, perhaps, into the esophagus
remains unresolved [16-18,22-24,28]. Several techniques have been introduced to detect "enterogastroesophageal" reflux. The most direct technique employs an indwelling esophageal tube to aspirate the esophageal contents and measure bile salts, bilirubin or pancreatic enzymes. Scintigraphic methods have been introduced employing ע labeled hepatocystic agents, usually derivatives of iminodiacetic acid (IDA), to mark bile salts so that their regurgitation into both the stomach and the esophagus can be identified. To date, this technique has been used mostly to detect and quantify enterogastric (bile) reflux in patients who have undergone peptic ulcer surgery. Recently, bilirubin and bile acid-sensing probes have been introduced to detect and quantitate enterogastric and enterogastroesophageal reflux. To date, little published data are available from studies using these new probes.
Determinants of physiological gastroesophageal reflux
The gastroesophageal pressure gradient is between 5 and 10 mmHg throughout much of the day. Therefore, if it were not for the antireflux barrier at the gastroesophageal junction, reflux would be continuous. Exceptions occur after swallowing or esophageal distension when primary or secondary peristaltic contractions, respectively, migrate aborally through the esophagus. Reflux could occur at times of increased intra-abdominal pressure during exercise, bending or straining to move the bowels. Intragastric pressures increase during the fed or postprandial period when antral contractions become more pronounced and regular. However, despite a pressure gradient conducive to reflux for most of the day, esophageal pH is below 4 for only one twentieth of the total time. Of course, this does not include nonacid reflux which would not be detected by intraesophageal pH monitoring.
The major antireflux factors are the gastroesophageal junction high pressure zone, constituted by the lower esophageal sphincter and the crural diaphragm, and the distal paraesophageal pressure. It is generally accepted that the lower esophageal sphincter contributes a tonic junction pressure of 12 to 26 mmHg above the intragastric pressure [1,2,4,26-28]; the crural diaphragm is mostly responsible for phasic changes in the high pressure zone but also, perhaps, for a component of the tonic pressure [3,19,20]. The high pressure zone pressure may be affected by a number of external factors. The most obvious are certain foods such as fat, xanthines (chocolate, cola), onions, peppermint and smoking which may decrease the resting LES pressure by a variety of mechanisms. When food-induced physiologic decreases in resting LES pressure are accompanied by increased postprandial antral contractility, this may theoretically predispose to physiologic reflux events. However, the most important factor in the production of physiological reflux is spontaneous (inappropriate or transient) LES relaxation which occurs independently of swallowing, mostly during the postprandial or fed period [7,8,21,29]. Dent and his colleagues have attributed more than 60% of physiological reflux episodes to these spontaneous LES relaxations. The mechanism for spontaneous LES relaxation may be an aberrant swallowing mechanism, which does not stimulate esophageal peristalsis, as evidenced by submental EMG recording. Another mechanism for physiologic reflux episodes may
be swallows associated with abnormal or incomplete esophageal peristalsis which may account for up to 10%. A minority of reflux episodes are due to repetitive swallows over a short time period leading to prolonged LES relaxation. It is of interest that physiologic increases in intra-abdominal pressure alone, which occur during exercise, do not usually cause gastroesophageal reflux because of adaptive pressure increases at the high pressure zone which more than compensate for the elevated intra-abdominal pressure [1,2]. These adaptive high pressure zone responses are mediated by cholinergic mechanisms.
Coordination between esophageal contractions, LES relaxation and increased postprandial gastric motility may be important in determining physiologic gastroesophageal reflux. However, to date, gastric contractions have not been measured in most studies of this phenomenon.
Esophageal motility is undoubtedly an important determinant of physiological esophageal acid exposure [6,10]. After some reflux episodes, a secondary esophageal contraction strips the refluxate from the body of the esophagus [13,15]. In addition, abnormal primary esophageal contractions have been reported during some physiologic reflux episodes. Equally important in limiting esophageal acid exposure is the saliva and its bicarbonate content which neutralizes the acid refluxate in a stepwise fashion . Fortunately, in normal subjects little reflux occurs in the recumbent position during sleep when swallowing and primary peristaltic contractions are reduced to one to three per hour and salivation is diminished.
To summarize, the factors which separate physiological from nonphysiological (pathogenic) gastroesophageal reflux are not clear. Most likely, the composition and volume of the refluxate, the contact time between the refluxate and the esophageal mucosa, and the defenses of the esophageal mucosa determine whether gastroesophageal reflux will be physiological or whether it will produce symptoms and complications.
3. Csendes A, Miranda M, Espinoza M et al. Parameter and location of the muscular gastroesophageal position or cardia in control subjects and in patients with reflux esophagitis or achalasia Scand J Gastroenterol 1981:16:951-956.